Dynamoelectric machines such as motors and generators are widely employed in industrial and commercial facilities. Many facilities operate several hundred or even thousands of such machines concurrently and these machines are often integral components of large interdependent processes. Accordingly, the machines are each depended upon for prolonged consistent operation whereby it is extremely advantageous to provide reliable failure prediction information. Of particular relevance in the present invention are bearing-related failures and, more particularly, failures related to lubrication problems in antifriction bearings. Diagnostic studies have consistently reflected that bearing-related failures are a substantial cause (about 42% of reported failures) of motor failures.
An antifriction bearing is designed to constrain rotary or linear motion while minimizing wear and other losses such as friction. Examples of this type of bearing are sleeve bearings, hydrodynamic bearings, and rolling element bearings. The most prevalent bearing type found on medium and low horsepower (e.g. fractional to 500 hp) motors are rolling element bearings such as ball bearings. To this end, typical antifriction bearings normally include a bearing housing defining an annular chamber and a plurality of rolling elements retained within the chamber. The bearing housing typically includes two annular components known as raceways, and more particularly an outer raceway and an inner raceway having interior surfaces which form the radial walls of the bearing chamber. (In the context of the present invention, “interior” corresponds to the relation of the surface relative to the chamber.) For example, the outer raceway may be mounted to a machine (e.g., a motor) and is intended to remain stationary relative thereto, while the inner raceway supports a rotating member (e.g., the motor's rotor or shaft).
The rolling elements may be either balls or rollers and the bearing may include one or more rows of such rolling elements. A cage is usually provided to retain the rolling elements in their correct relative positions so that they do not touch one another and to provide some guidance for the rolling elements. Also a lubricating fluid, such as oil or grease, is contained within the bearing chamber to reduce the friction between the components and also assist in the dissipation of heat. The top and bottom (or axial) ends of the chamber are sealed by the mounting structure or by sealing covers to maintain the lubricating fluid within the bearing chamber and/or keep dirt or other contaminants out. An antifriction bearing may include a circulating system to inject and/or drain lubricating fluid into the bearing chamber.
The loss of lubricating effectiveness will result in accelerated wear of the bearing elements, additional heat generation due to frictional effects, higher levels of vibration and potential impact loading due to metal-to-metal contact, and accelerated degradation of lubricant health due to higher levels of temperature, metal particulate contamination, and higher loading/shear levels.
Needless to say, the health of lubrication is a significant factor in the overall operation of an antifriction bearing. Accordingly, it is essential that the lubrication of an antifriction bearing be properly provided, protected, and maintained. Initially, it is important that the correct lubricating fluid be provided for the antifriction bearing. Also, it is critical that an adequate amount of lubricating fluid be maintained in the bearing. Likewise, it is crucial that contaminants (such as water, rust, and other contaminations) not contaminate the lubricating fluid. Moreover, when the lubricating fluid is continuously exposed to elevated temperatures, accelerated speeds, high stress/loads, and an oxidizing environment, the lubricating fluid will inevitably deteriorate and lose its lubricating effectiveness.
Also, when machinery is re-lubricated by applying additional lubricant, the addition of a different, incompatible lubricant will result in considerably diminished lubricating effectiveness. The result may be accelerated bearing wear beyond what would occur if no additional lube was added. The loss of bearing lubricant due to a seal failure is important to detect to prevent accelerated bearing wear and to avoid a dry running condition. It is also important to detect the loss of bearing lubricant in critical manufacturing processes such as pharmaceutical, medical products, and food products manufacturing. Loss of lubricant could result in a contaminated product or worse a contaminated product which remains undetected before reaching the consumer.
Lubrication-related problems tend to be insidious. There is often only a minor degradation of the lubricating fluid at the beginning. However, continued operation of the machine results in even greater heat generation and accelerated degradation of the lubricating fluid. Left untreated, the bearing will eventually fail leading to substantial machinery damage. In short, the continued operation of a degraded bearing will generally destroy machinery beyond just the bearing and repair costs can be extremely high, not to mention the catastrophic and potentially unsafe conditions such a failure creates. Unfortunately, many lubrication-related problems are only recognized after irreparable destruction has occurred to not only the bearing, but the machine itself. For example, some lubricant problems eventually result in a bearing seizing up and the continued rotary motion destroying the rotating shaft or the bearing mounting. Alternatively uncontrolled vibration could occur, resulting in destruction of machinery and buildings.
The previous discussion presented bearings and lubrication issues from the standpoint of motor-mounted bearings. The problems identified and the need for lubricant health information applies to bearings found in a wide range of machinery, including machinery connected to a motor (driven equipment), land-based vehicles, shipboard systems and aircraft systems. This includes bearings in internal combustion engines, engine accessories, gears and gear systems, wheels, linear slides, conveyors, rollers, and pillow blocks for example.
Accordingly, an early diagnosis and cure of lubrication-related problems can be extremely beneficial in reducing machine down-time, repair cost, inconvenience, and even hazardous situations. For this reason, conservative lubricant changing schedules (where the lubricating fluid is changed well before any degradation is expected to occur) are sometimes well worth what may be viewed as wasted labor and wasted lubricating fluid and un-necessary machinery downtime if needed. Other times, however, the cost and labor associated with replacing adequate functioning lubricating fluid cannot be justified. Additionally, the more frequent the changes, the higher the possibility that the wrong lubricating fluid will be provided and/or other changing mistakes will be made such as over lubricating equipment. More importantly, each lubrication situation seems to be relatively unique in view of the almost countless factors that can contribute to lubrication degradation. As such, in many situations, a lubricating fluid will reach at least the initial stages of breakdown or contamination well before even a conservative scheduled change.
The potential damage associated with inadequate bearing lubrication and the uniqueness of each lubrication situation has led many industries to adopt programs of periodic monitoring and analyzing of the lubricating fluid. In some programs, for example, the condition of the lubrication is determined by measuring a dielectric constant change in the lubricating fluid. In other programs, for example, the condition of the lubrication is instead determined by recording historical thermal readings. Unfortunately, these programs measure only a single parameter, such as temperature over time, require the use of the same lubricating fluid, and/or assume no machinery malfunctions between measurements. Furthermore, extensive lubrication monitoring and analysis techniques are not performed in situ and typically involve extracting a sample of the lubricating fluid and then analyzing this sample using laboratory equipment. As such, these sampling techniques only provide a narrow, periodic view of lubrication quality and/or health whereby accurate lubrication health assessment and lifetime prediction is virtually impossible to achieve. Moreover, the manpower required to extract the lubricant samples necessarily limits the frequency of sampling, not to mention the introduction of human error into the extraction process. In some situations, lubricating oil may be extracted from machinery and put in glass bottles in front of a light source. A visual inspection is made after the material had settled.
In view of the above, analysis of fluids including lubricating fluids is an extremely important area and rapidly growing in importance (e.g., machinery, safety, environmental, and so forth). There is a significant expenditure of dollars for outside lab analysis of fluids and also for staff time and for on-site factory-resident labs. Also, there are a variety of lube analysis techniques that include lab analysis methods (e.g., titration methods) and sensor-based methods (e.g., pH sensors, H2O sensors, dielectric sensors). Lab analysis techniques, however, are limited due to the time delay before a lube analysis is available, possible contamination of the samples extracted for analysis, the cost required for the analysis, and the difficulty in determining what corrective action is needed and when.
Although lube sensors offer improvement over how machines were maintained in the past, other problems are still encountered. For example, maintenance engineers generally still must go to the machinery and lubricate equipment periodically based on equipment usage and lubrication health. In addition, many pieces of rotating machinery and associated grease fittings are difficult to reach, whereby other problems relate to accelerated failure due to over lubricating equipment and by employing the wrong type of lubrication, for example.